The study's purpose is to investigate the relationship between treatment variables and rehabilitation outcome in heroin addicts in a manner that allows the identification of causal links between treatment and rehabilitation. Three treatment modalities involving specific skill training will be employed: family adjustment skills, vocational/school skills, and interpersonal problem-solving skills. There will also be two contract groups: one which receives no special training, and a second which receives attention similar to that received by the three experimental groups, but devoid of any skill training. The rehabilitation charge variables will include both traditional outcome measures, and indices of intermediate change in attitudes, values, and self-concept.